FDA 510(k) Search - By Innolitics (SaMD and AI Experts)

The NeuViz Prime Multi-Slice CT Scanner System can be used as a whole body computed tomography X-ray system featuring a continuously rotating X-ray tube and detector array. The acquired X-RAY transmission data is reconstructed by computer into cross-sectional images of the body from either the same axial plane taken at different angles or spiral planes taken at different angles.

Device Description

The NeuViz Prime Multi-Slice CT Scanner System is composed of a gantry, a patient couch, an operator console and includes image acquisition hardware and software, and associated accessories. It is designed to be a head and whole body X-ray computed tomography scanner which features a continuously rotating tube-detector system and functions according to the fan beam principle. The system provides the filter back-projection (FBP) and iterative reconstruction algorithm(ClearView cleared in K133373) to reconstruct images. The end user can choose to apply either ClearView or the FBP to the acquired raw data. The system software is an interactive program used for X-ray scan control, image reconstruction, and image archive/evaluation. It provides the following digital image processing and visualization tools:

· Support following scan speed: 0.259s(option), 0.32s(option), 0.374s(option), 0.4s(option), 0.5s, 0.6s, 0.8s, 1.0s, 1.5s, 2.0s.
. Surview scan
Dual surview
Spiral scan
Axial scan
Image reconstruction
Plan scan
Patient information management
Patient information registration
Protocol selection
O-Dose
· Bolus tracking
SAS
Home
Film
Report
2D
MPR
3D
VE(Virtual Endoscopy)
Vessel Analysis
Dicom Viewer
Bar code Reader
Dual Monitor
CCT Scan
ClearView
iHD
Cardiac Scan
Dual Energy Scan and Reconstruction ●
Dental Analysis
Virtual Colonoscopy
Brain Perfusion
Body Perfusion
· Lung Nodule Analysis
Lung Density Analysis
· Coronary Analysis
Cardiac Calcium Scoring .
Cardiac Function Analysis
Cardiac Viewer
Fat Analysis
CTDSA
Tumor Assessment
· Preprocessing function
AVW.Cloud
· Prism Viewer

AI/ML Overview

The provided document is a 510(k) summary for the NeuViz Prime Multi-Slice CT Scanner System. It describes the device's characteristics and its substantial equivalence to a predicate device but does not contain acceptance criteria for specific performance metrics or detailed results of a study designed to prove the device meets those criteria, especially in an AI context. The document focuses on showing non-inferiority to an existing device rather than meeting specific quantifiable performance targets with clinical evidence in the format you requested for an AI/ML device.

However, I can extract information related to the device's performance characteristics, safety, and the "clinical testing" that was performed, even if it doesn't align perfectly with the AI/ML-focused questions.

Here's an attempt to answer your questions based on the provided text, while acknowledging the limitations for AI-specific criteria:

1. A table of acceptance criteria and the reported device performance

The document does not provide a table of explicit acceptance criteria with numerical targets for clinical performance (e.g., sensitivity, specificity, accuracy for a specific disease detection task). Instead, it states that "the subject device performs as intended" and "NeuViz Prime can be used as defined in its clinical workflow and intended use," and that "The Results indicated that the images were of diagnostic quality."

For image quality metrics, it lists:

CT number accuracy and uniformity
MTF (Modulation Transfer Function)
Noise
Slice sensitivity profiles
CTDI (Computed Tomography Dose Index)

The document doesn't report specific numerical acceptance criteria for these image quality metrics, nor does it provide the measured performance values for them. It only states that "The result of all conducted testing was found acceptable to support the claim of substantial equivalence."

It does provide CTDI Dose values for the subject device compared to the predicate:

Acceptance Criteria (Implied by Comparison)	Reported Device Performance (NeuViz Prime)	Predicate Device (NeuViz 128)	Comments
CTDI Dose (Head)	14.2 mGy/100mAs	13.0 mGy/100mAs	Approximately 10% higher due to beam filter and wedge material differences.
CTDI Dose (Body)	7.2 mGy/100mAs	6.5 mGy/100 mAs	Approximately 10% higher due to beam filter and wedge material differences.
Image Quality	Images were of diagnostic quality	N/A (implied similar)	Based on evaluation by a qualified radiologist using a 5-point Likert scale.
Functionality	Performs as intended	N/A	Verified through functional, smoke, and regression tests, adhering to software lifecycle processes and addressing potential defects.

2. Sample size used for the test set and the data provenance

The "clinical testing" involved an "image evaluation" where "sample images were provided to show the performance of the system in presence of implants." This suggests a test set composed of image data.

Sample Size: Not explicitly stated. The document refers to "images of the brain, chest, abdomen and spine/extremities of the body area," implying multiple images, but no specific count is given.
Data Provenance: Not explicitly stated. It is likely retrospective data as it describes an "image evaluation" of existing images rather than a prospective clinical trial. The location of data origin (e.g., country) is not mentioned.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

Number of Experts: One.
Qualifications of Experts: "A qualified radiologist." No specific experience level (e.g., "10 years of experience") is provided.

4. Adjudication method for the test set

Adjudication Method: Not explicitly an adjudication method in the sense of multiple readers reaching consensus. The images were "scored using a 5 point Likert scale by a qualified radiologist." This implies a single-reader assessment rather than a consensus or adjudicated ground truth process.

5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

MRMC Study: No, a multi-reader multi-case comparative effectiveness study was not reported. The provided document describes a CT scanner system, not an AI-assisted diagnostic tool for which human reader improvement would be typically measured. The "clinical testing" described was an image quality assessment by a single radiologist.

6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done

Standalone Performance: Not applicable in the context of this submission. The NeuViz Prime is a CT scanner, a hardware device that generates images. While it has reconstruction algorithms (FBP, ClearView, iHD) and image processing tools (e.g., Lung Nodule Analysis, Cardiac Calcium Scoring), the submission focuses on the overall performance of the imaging system and its substantial equivalence to a predicate device, not on the standalone performance of an AI algorithm intended for diagnostic interpretation. It does mention "The main algorithm of Prism Viewer Application is identifying of substances and calculating of dual energy images," but no standalone performance metrics are provided for this.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

Type of Ground Truth: The document refers to an "image evaluation" where images were "scored using a 5 point Likert scale by a qualified radiologist." This points to expert opinion/scoring as the basis for evaluating "diagnostic quality." It does not mention pathology, outcomes data, or a consensus of multiple experts.

8. The sample size for the training set

Training Set Sample Size: Not applicable/not provided. This document describes the clearance of a CT scanner system, not an AI/ML algorithm that would typically have a distinct training set. While the system's reconstruction algorithms (like ClearView, iHD) would have been developed and "trained" or optimized during their creation, this document does not provide details on their specific training sets.

9. How the ground truth for the training set was established

Ground Truth for Training Set: Not applicable/not provided. Similar to point 8, the document does not discuss the training of AI/ML models. If "ClearView" or other advanced algorithms involved machine learning at their core, the method for establishing their training ground truth is not detailed in this 510(k) summary.

Summary

Regulation Number and Section

§ 892.1750 Computed tomography x-ray system.

(a)
Identification. A computed tomography x-ray system is a diagnostic x-ray system intended to produce cross-sectional images of the body by computer reconstruction of x-ray transmission data from the same axial plane taken at different angles. This generic type of device may include signal analysis and display equipment, patient and equipment supports, component parts, and accessories.(b)
Classification. Class II.